Skip to main content
NCBI home page
Journal of Bacteriology logoLink to Journal of Bacteriology
. 1998 Apr;180(7):1887–1894. doi: 10.1128/jb.180.7.1887-1894.1998

A Small Protein (Ags1p) and the Pho80p-Pho85p Kinase Complex Contribute to Aminoglycoside Antibiotic Resistance of the Yeast Saccharomyces cerevisiae

Stephan Wickert 1, Markus Finck 1, Britta Herz 1, Joachim F Ernst 1,*
PMCID: PMC107104  PMID: 9537389

Abstract

We identified the AGS1 and AGS3 genes by their ability to partially complement an ags mutant (RC1707) which is supersensitive to various aminoglycoside antibiotics (J. F. Ernst and R. K. Chan, J. Bacteriol. 163:8–14, 1985). AGS1 is located in proximity to the centromere of chromosome III and encodes a small protein of 88 amino acids. The size of the AGS1 transcript, which in wild-type cells is 1 kb, is reduced to 0.75 kb in mutant RC1707. Disruption of AGS1 rendered strains supersensitive to hygromycin B and increased their resistance to vanadate. In addition, ags1Δ strains underglycosylated invertase but had normal carboxypeptidase Y glycosylation, suggesting that Ags1p is required for the elaboration of outer N-glycosyl chains. AGS3 was found to be identical to PHO80 (TUP7), which encodes a cyclin activating the Pho85p protein kinase. Deletion of either PHO80 or PHO85 led to aminoglycoside supersensitivity; pho80Δ ags1Δ strains showed an enhanced-sensitivity phenotype compared to single mutants. pho80 and pho85 mutants were rendered resistant by deletion of PHO4, indicating that activation of the Pho4p transcription factor is required for increased aminoglycoside sensitivity. Thus, both the Pho80p-Pho85p kinase complex (by Pho4p phosphorylation) and a novel component of the N glycosylation pathway contribute to basal levels of aminoglycoside resistance in Saccharomyces cerevisiae.

In prokaryotic and eukaryotic cells, aminoglycoside antibiotics promote mistranslation by specific interactions with small rRNA (2, 8, 14, 25, 31). At low concentrations of aminoglycosides, mistranslation leads to phenotypic suppression of mutant phenotypes in Saccharomyces cerevisiae (37, 44). Additional inhibitory activities of aminoglycosides are caused by binding to specific mRNA sequences and include the block of group I intron splicing (46) and the inhibition of binding of the human immunodeficiency virus Rev protein to its cognate RNA sequences (47).

Mutations lowering or increasing the susceptibility to aminoglycosides in yeast are known. Mutations in PMA1, encoding a plasma-membrane ATPase (38), as well as mutations in the mitochondrial and cytoplasmic rRNAs (8, 14, 25) confer aminoglycoside resistance. Resistance has also been generated experimentally by overexpression of the yeast NEO1 gene, encoding a homolog of a cytoplasmic ATPase that is involved in ribosome assembly (39), or by expression of bacterial genes encoding aminoglycoside-modifying enzymes (e.g., aph or hph genes encoding bacterial phosphotransferases [12]), which can serve as dominant selectable markers in expression plasmids (18, 19). Aminoglycoside supersensitivity can arise because of mutations causing defects in N glycosylation (3, 13) and O glycosylation (45a). crl mutants are supersensitive to hygromycin B and other aminoglycosides but are resistant to cycloheximide (28). Similarly, pdr1 mutants are sensitive to some aminoglycosides but resistant to other compounds and show some respiratory defects (40). An increase in aminoglycoside sensitivity can also arise because of the presence of certain omnipotent translational suppressors (26); mutations in MOF4, whose gene product is involved in reading frame maintenance, increase the sensitivity to paromomycin (10). Furthermore, specific mutations in 18S rRNA significantly enhance the susceptibility of yeast to the aminoglycoside streptomycin (8).

We previously described ags mutant strains of S. cerevisiae, which show enhanced sensitivity to aminoglycosides, including G418, hygromycin B, destomycin A, gentamicin X2, apramycin, kanamycin B, lividomycin A, neamine, neomyin, paromomycin, and tobramycin (15). The ags phenotype was not associated with any of the sensitivities and additional phenotypes caused by previously described mutations. Genetic analyses suggested that the aminoglycoside supersensitivity of ags strains was caused by mutations in three recessive genes, one of which (AGS1) was found to map close to the centromere of chromosome III. The second gene involved in supersensitivity (AGS2) appeared to be linked to AGS1 on chromosome III, while the third gene, designated AGS3, was unlinked to AGS1 and situated on an undefined chromosome. Here we describe the cloning and characterization of two genes complementing the supersensitivity of the ags strains. We show that a small unknown protein (Ags1p) and known proteins (Pho80p, Pho85p, and Pho4p) contribute to a basal level of resistance to aminoglycosides in wild-type yeast cells.

MATERIALS AND METHODS

Strains and growth conditions.

The strains used are listed in Table 1. Cells were grown in complex YPD or synthetic SD medium (43). Sensitivities to antibiotics were determined on YPD plates containing a linear gradient of 0 to 25 μg of hygromycin B per ml or 0 to 300 μg of G418 (Geneticin; Sigma) per ml. The ags phenotype was routinely tested on YPD plates containing 1, 5, and 10 μg of hygromycin B per ml; ags mutants were inhibited by hygromycin B at 5 and 10 μg/ml, while wild-type strains were able to grow at these hygromycin B concentrations. ags1 pho80 double mutants were obtained as segregants of a cross of W10-2B and VG51.

TABLE 1.

Yeast strains

Name Genotype Reference or source
BJ1991 MATα pep4-3 prb1-1122 ura3-52 leu2 trp1 23
RC1705 MATa ura3-52 trp1 prb1-1122 15
RC1707 MATa ura3-52 trp1 tup7 (pho80) pep4-3 prb1-1122 ags 15
CEN-PK141 MATa/MATα leu2-3,112/LEU2 ura3-52/URA3 trp1-298/TRP1 his3-Δ1/HIS3 MAL2-8c/ MAL2-8cSUC2/SUC2 7
W10-2A MATα ura3-52 ags1Δ::loxP-kanMX-loxP This work
W10-2B MATa his3-Δ1 ura3-52 leu2,3-112 ags1Δ::loxP-kanMX-loxP This work
W10-2C MATα his3-Δ1 trp1-289 This work
W10-2D MATα trp1-289 leu2-3,112 This work
WK4a-7A MATα his3 ura3 pho80-Δ1 ags1Δ::loxP-kanMX-loxP This work
WK4a-7C As WK4a-7A but MATa This work
VG51 MATα his3 trp1-Δ1 ura3-167 leu2 pho80-Δ1 17
BY263 MATa trp1-Δ63 ura3-52 lys2-801amade2-107ochis3-Δ200 leu2-Δ1 B. Andrews
BY391 As BY263 but pho85Δ LEU2 B. Andrews
BY490 As BY263 but pho80Δ HIS3 B. Andrews
SH8245 (NBW7) MATa pho3-1 leu2-3,112 ura3-1,2 trp1-289 his3-532 ade2 can1 S. Harashima
SH8249 As SH8245 but pho80::HIS3 S. Harashima
SH8405 As SH8245 but pho85::ADE2 S. Harashima
SH8455 As SH8245 but pho80::HIS3 pho4::HIS3 S. Harashima
SH8456 As SH8245 but pho85::ADE2 pho4::HIS3 S. Harashima
G2-10 MATα ura3-52 lys2-801amade2-101octrp1-Δ1 his3-Δ200 leu2-Δ1 1
G2-11 As G2-10 but gda1::LEU2 1
PRY98 MATα alg5-1 ade2-101 ura3-2 21
Open in a new tab

Genes complementing the ags phenotype.

The ags strain RC1707 (15) was transformed with an S. cerevisiae genomic library constructed either in the replicating vector YRp7 (33) or in the centromeric vector YCp50 (41), using the lithium acetate transformation protocol (22). Transformants grown selectively for 2 days at 30°C on supplemented SD medium were replica plated to YPD medium containing 20 μg of G418 per ml. Resistant transformants were retested to confirm enhanced resistance to hygromycin B and G418. Furthermore, chromosomal DNA of the resistant transformants was prepared, and plasmids were recovered by transformation into Escherichia coli (43). Retransformation of the obtained plasmids into RC1707 confirmed that antibiotic resistance was plasmid borne. Of 73,400 transformants obtained with the YCp50 library, 7 transformants carried a plasmid conferring resistance; of 33,200 transformants obtained with the YRp7 library, 13 were identified as conferring resistance.

Plasmids.

Twelve of 13 G418- or hygromycin B-resistant transformants obtained with the YRp7 genomic library carried a plasmid with an insert of 8.5 kb, which was designated pA8. In the case of the YCp50 library, all seven resistant transformants carried a plasmid with a genomic insert of about 12 kb; this plasmid was designated pB12. Subclones of the genomic inserts were constructed in the centromeric vector YCplac22 (16). The YCplac22 subclone carrying the 4.7-kb EcoRI-SalI fragment derived from pA8, which conferred identical antibiotic resistance, was designated YCplac22-4.7. Similarly, the YCplac22 subclone carrying the 8-kb EcoRI fragment derived from pB12 conferred the same resistance as pB12 and was designated pB6A. (All constructed subclones are summarized in Fig. 2.) pMD1.9 was constructed from pB6A by inserting the 1.6-kb SacI-SpeI AGS1 fragment into YCplac22 between the SacI and XbaI sites of YCplac22. pSW17 was constructed by insertion of the 4.7-kb EcoRI-SalI fragment (YCplac22-4.7) carrying PHO80 into YCplac33 (16). pSW18 was constructed by insertion of the 6.5-kb EcoRI fragment (YCplac22-6A) carrying AGS1 into YCplac33 (16).

FIG. 2.

FIG. 2

Open in a new tab

Complementation of aminoglycoside supersensitivity of strain RC1707 by genomic clones. The extent of genomic inserts in the indicated plasmid is shown by the black bars. The ability of the clones to confer resistance to G418 is indicated by +, −, or +/−; the complementing activity of the original clone is designated +. A restriction map of the genomic region along with the coding region of AGS1 (top) or PHO80 (bottom) (open arrows) is shown. B, BamHI; K, KpnI; X, XbaI; R, EcoRI; S, SalI; Sc, SacI; Se, SpeI; Sh, SphI. The EcoRI sites flanking the genomic insert in pB6A do not occur in the published sequence of chromosome III; therefore, these sites are in parentheses.

Gene disruptions.

Gene disruptions were performed with strain CEN-PK141 by a PCR method (19); strain CEN-PK141 is isogenic to CEN-PK2 (7) except that it is a MATa/MATα diploid and is heterozygous for the auxotrophic markers. To delete the entire coding region of AGS1, the primers AGS-L (5′-ATG GAC ATG GAC AAC ACG GAT ATC TCC CCA ACC AAC CAT CCA GCT GAA GCT TCG TAC GC-3′) and AGS-R (5′-CTA TTT GCG GTT CAG GAC GTC TAT CTG TGC GTT TAG CGA GGC ATA GGC CAC TAG TGG ATC TG-3′) were used first to amplify the loxP-kanMX-loxP (Kanr) module (italic indicates homology to AGS1; boldface indicates homology to loxP) on plasmid pUG6 (19). The PCR product was transformed into CEN-PK141, the resulting diploid transformant was sporulated, and the deletion was confirmed by Southern blotting with the 3.1-kb NcoI-PstI fragment downstream of AGS1 as a probe. The disrupted diploid was sporulated, and haploid ags1Δ segregants were identified by their G418 resistance.

Other procedures.

Total RNA was prepared as described previously (42), separated by denaturing gel electrophoresis, transferred to nylon filters, and probed with a 32P-labelled DNA fragment carrying AGS1 (1-kb BamHI-SalI fragment of YCplac22-1) according to standard procedures. Invertase (encoded by SUC2) was analyzed by activity staining following native acrylamide gel electrophoresis (32); alternatively, proteins in cell extracts were denatured and separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) (5% acrylamide), and invertase was detected by Western blotting with a polyclonal rabbit anti-invertase antibody (1:1,000) followed by goat horseradish peroxidase-coupled anti-rabbit immunoglobulin G (1:5,000). Likewise, carboxypeptidase Y (CPY) in cellular proteins was detected after SDS-PAGE (7.5% acrylamide) with a monoclonal anti-CPY antibody (1:500) (Molecular Probes) followed by goat horseradish peroxidase-coupled anti-mouse immunoglobulin G antibody (1:10,000) (Dianova) (48).

RESULTS

Genomic clones complementing the Ags− phenotype.

We previously characterized strain RC1707, which harbors at least three mutations leading to increased sensitivities to several aminoglycoside antibiotics (15). To identify genes involved in basal aminoglycoside resistance of wild-type yeast strains, we transformed RC1707 with genomic libraries (33, 41) and selected transformants able to grow at 10 to 20 μg of G418 per ml, concentrations at which wild-type strains, but not ags strains, are able to grow. Resistant transformants were analyzed further by allowing plasmid loss during nonselective growth in YPD medium; in all cases the plasmid-free derivative strains were as antibiotic sensitive as the parental strain RC1707. Furthermore, plasmids carried by the resistant transformants were recovered by transformation of chromosomal DNA into E. coli; RC1707 transformants carrying the reisolated plasmid were as resistant as the original transformants. The latter experiments indicated that in all cases, aminoglycoside resistance of the isolated transformants was plasmid borne.

By these criteria, 13 plasmids conferring resistance were identified in a genomic library constructed in the multicopy vector YRp7 (33); in addition, 7 plasmids conferring resistance were isolated from a genomic library constructed in the centromeric vector YCp50 (41). Restriction analyses revealed that 12 of the 13 isolated YRp7 derivatives were identical and carried a genomic insert of 8.5 kb; a representative plasmid was designated pA8. Similarly, restriction analyses of the YCp50 clones showed that they contained an identical genomic insert of about 12 kb; a representative plasmid was designated pB12. RC1707 transformants carrying plasmid pA8 or pB12 showed increased G418 resistance compared to the untransformed RC1707 strain; these transformants, however, were less resistant than the wild-type strain (Fig. 1). An identical pattern of sensitivity and resistance of strains was obtained by using hygromycin B (data not shown).

FIG. 1.

FIG. 1

Open in a new tab

Sensitivity of yeast strains to G418. The orientation of the G418 gradient is indicated at the top. Growth of wild-type strain ER110-6B and growth of ags mutant RC1707 untransformed or transformed with genomic clone pB12 or pA8 are compared.

Subcloning experiments.

To delimit the genomic region in pA8 and pB12 which was responsible for complementation of the Ags− phenotype, we constructed various plasmid subclones and tested their abilities to confer G418 and hygromycin B resistance in transformants of strain RC1707 (Fig. 2).

A YRp7 derivative carrying a 4.7-kb EcoRI-SalI genomic fragment derived from pA8 (pA84) was found to convey levels of resistance similar to those with pA8. To explore if resistance was due to the high copy number of pA84, we transferred the 4.7-kb fragment to the centromeric vector YCplac22, resulting in plasmid YCplac-4.7 (Fig. 2). RC1707 transformants carrying YCplac-4.7 were as resistant as transformants carrying pA8 and pA84, indicating that complementation by the genomic 4.7-kb fragment can occur at low plasmid copy numbers. From the activities of additional subclones, we concluded that a 2.7-kb KpnI-SalI genomic fragment contained in YCplac22-2.7, which is derived from A8, is sufficient to confer G418 and hygromycin B resistance (Fig. 2).

A 6-kb genomic EcoRI fragment, which is contained in pB12, was subcloned into YCplac22 (resulting in plasmid pB6A); resistances of transformants carrying pB12 or pB6A were identical. The genomic insert of pB6A was subcloned further, and a subclone containing a 1.9-kb SacI-SpeI fragment (pMD-1.9) was found to confer the same resistance levels as pB6A and pB12 (Fig. 2). A subclone containing an even smaller, 1-kb BamHI-SalI fragment (YCplac22-1) was able to complement the Ags− phenotype partially.

Sequences of genomic clones.

The sequence of the 1-kb BamHI-SalI fragment of YCplac22-1 and the terminal sequences of the 2.7-kb KpnI-SalI fragment of YCplac22-2.7 were determined.

Computer analyses revealed that the sequence of the 1-kb BamHI-SalI fragment of YCplac22-1 was identical to that of a segment on chromosome III between the centromere and the LEU2 gene. A single complete open reading frame was detected on this fragment, extending from bp 105841 to 105578 of chromosome III; this reading frame has the potential to encode a protein of 88 amino acids (Fig. 3). We had previously mapped a gene, designated AGS1, involved in aminoglycoside resistance close to the centromere of chromosome III (15). Because of the complementation results, as well as the altered transcript size in RC1707 and the phenotype of the disruptant strain (see below), we designate the identified open reading frame AGS1. Ags1p did not show significant homologies to any other protein in the databases. The Ags1 protein is rich in hydrophobic and acidic residues but does not contain known sorting signals or transmembrane regions (Fig. 3A). Computer analyses (http://expasy.hcuge.ch/sprot/prosite.html) predict that the N-terminal 20 amino acids are in a β-turn or coil configuration, while residues 20 to 88 adopt a β-sheet conformation.

FIG. 3.

FIG. 3

Open in a new tab

Translational product and transcript of AGS1. (A) Theoretical protein encoded by AGS1. Hydrophobic residues are shaded, acid residues are marked by closed circles, and basic residues are marked by open circles. (B) AGS1 and ags1 transcripts. Total RNAs of strain RC1707 (lane 1), strain RC1707(pB6A) (lane 2), and the wild-type strain BJ1991 (lane 3) were probed with a 1-kb BamHI-SalI probe of YCplac22-1 carrying AGS1. The migrations of 16S and 25S rRNAs are indicated.

Computer analyses of the terminal sequences of the genomic fragment contained in YCplac22-2.7 revealed that it corresponds to a fragment in close vicinity to the centromere of chromosome XV. Only a single open reading frame (YOL001W) was detected in this region, extending from bp 325249 to 326130 of chromosome XV. YOL001W corresponds to the PHO80 gene, which encodes a cyclin activating the Pho85p kinase involved in phosphate regulation (36). We initially referred to this gene as AGS3, because it is located on a different chromosome than AGS1 (15), but because of its discovered identity to PHO80, we refer to it here by this established name.

AGS1 transcript.

To characterize the AGS1 transcript, we isolated total RNA from a wild-type strain (BJ1991), the ags strain RC1707, and a transformant of RC1707 carrying the genomic clone pB6A. Northern blots were prepared and probed with the SalI-BamHI fragment carrying AGS1 (Fig. 3B).

In the AGS1 wild-type strain BJ1991, a single transcript of about 1 kb was obtained, indicating that AGS1 is expressed (Fig. 3B, lane 3). From the size of its coding region, we deduce that the 5′ and 3′ untranslated regions of the AGS1 transcript amount to about 0.7 kb. In contrast, the ags strain RC1707 contained a shortened transcript of about 0.75 kb (Fig. 3B, lane 1), suggesting that an extensive deletion has occurred in the AGS1 gene of this strain. As expected, the RC1707 transformant carrying the genomic clone pB6A expresses the deleted ags1 transcript, as well as the wild-type AGS1 mRNA (Fig. 3B, lane 2).

The transcript of the PHO80 (AGS3) gene has been characterized previously (27).

Construction and phenotype of ags1Δ strains.

To verify the function of AGS1, its coding region was replaced by a loxP-kanMX-loxP module in the diploid strain CEN-PK141 (see Materials and Methods). Replacement of the AGS1 coding region by the loxP-kanMX-loxP module introduces a new PstI site at the location of the deleted AGS1 (Fig. 4, top). Thus, on Southern blots with the 3.1-kb PstI-NcoI fragment downstream of AGS1 as a probe, DNA of wild-type cells is expected to contain a 7.7-kb PstI fragment, while the size of this fragment is reduced to 3.8 kb in ags1Δ::loxP-kanMX-loxP strains. Two G418-resistant diploids containing both the 7.7- and 3.8-kb PstI fragments were obtained (Fig. 4, bottom, lanes 1 and 2). The diploids were sporulated, tetrads were dissected, and haploid segregants were analyzed.

FIG. 4.

FIG. 4

Open in a new tab

AGS1 disruption in the diploid CEN-PK141. A schematic of the disruption of AGS1 (open arrow) by the loxP-kanMX-loxP module is shown at the top. N, NcoI; P, PstI. A Southern blot demonstrating the disruption of one of the two AGS1 alleles is shown at the bottom. Genomic DNAs of two disruption strains (lanes 1 and 2) and the original strain CEN-PK141 (lane 3) were cut with PstI and analyzed by Southern blotting with the 1-kb BamHI-SalI AGS1 fragment as a probe. The migrations of the wild-type AGS1 fragment (7.7 kb), the ags1Δ fragment (3.8 kb), and standard fragments (4.3 and 5.1 kb) are as indicated.

As expected, G418 resistance due to the ags1Δ::loxP-kanMX-loxP disruption segregated 2:2 with G418 sensitivity (AGS1). Interestingly, the ags1Δ haploids were found to be supersensitive to hygromycin B at a concentration of 10 μg/ml. This phenotype was verified on hygromycin B gradient plates, as shown for the two haploid ags1Δ segregants W10-2A and W10-2B (Fig. 5A). Thus, deletion of AGS1 reproduces the aminoglycoside supersensitivity phenotype observed originally in strain RC1707. Reintroduction of the intact AGS1 gene into W10-2B by transformation with plasmid pSW18 reconstituted the wild-type resistance level, while a control plasmid (YCplac33) did not complement the ags1Δ defect (Fig. 5A).

FIG. 5.

FIG. 5

Open in a new tab

Hygromycin B resistance of yeast strains. The indicated strains were streaked on a linear gradient of hygromycin B (0 to 120 μg/ml) in YPD medium and grown for 2 days at 30°C. The effects of the ags1 mutation (A), the pho80, pho85, and pho4 mutations (B) and a combination of ags1 and pho80 (C) were tested.

Besides aminoglycoside supersensitivity, ags1 mutants did not show any other phenotype compared to AGS1 strains; growth rates on media containing glucose or sucrose as carbon sources, as well as the microscopic appearances of cells, were identical in ags1Δ and AGS1 wild-type strains.

Phenotypes of pho80 (ags3) and pho85 strains.

As shown above, we had identified the PHO80 gene on one type of genomic clone complementing the supersensitivity phenotype of strain RC1707. To test if PHO80 is involved in aminoglycoside resistance of yeast, we compared the hygromycin B sensitivities of a pho80 mutant (SH8249) and an isogenic wild-type strain (SH8245). Furthermore, since PHO80 encodes a cyclin activating the Pho85p protein kinase, we also examined whether aminoglycoside resistance was affected in an isogenic pho85 mutant (SH8405). As shown in Fig. 5B, both pho80 and pho85 mutants were significantly more sensitive than the wild type to hygromycin B; an increased sensitivity of both mutants to G418 was also observed (data not shown). An identical result was obtained with a different isogenic set of strains (BY391 [pho85] and BY490 [pho80]) derived from strain BY263 (30) and with strain VG51 (pho80). These results suggested that the activity of the Pho80p-Pho85p complex is required to maintain wild-type levels of resistance against aminoglycosides.

One of the functions of the Pho80p-Pho85p complex is to phosphorylate the Pho4 transcription factor and thereby to inactivate it under high-phosphate conditions (36). To examine whether an unrefrained activity of Pho4p was responsible for the increased sensitivity of the pho80 and pho85 mutants, we tested isogenic pho80 pho4 and pho85 pho4 double mutants for their sensitivity to hygromycin B and G418. In such double mutants, wild-type resistance levels were restored (Fig. 5B). These results suggested that an elevated activity of Pho4p is responsible for an increase in sensitivity to aminoglycoside antibiotics.

To test if ags1 and pho mutations have an additive effect in increasing aminoglycoside sensitivity, we crossed strain W10-2B (ags1Δ) with strain VG51 (pho80) and identified segregants. Among 12 tetrads, 2 tetrads (WK4a-7 and WK4b-5) which contained two wild-type spores and two ags1 pho80 double mutant spores were identified. Among such double mutants, in each tetrad, one segregant (WK4a-7A or WK4b-5A) indeed was more sensitive than the single mutants, while a second segregant (WK4a-7C or WK4b-5C) showed no enhanced sensitivity compared to the pho80 mutant (Fig. 5C). Thus, unknown genes present in the genetic background of the parental strains appear to influence the aminoglycoside sensitivity of the double mutants.

ags1 mutants have a defect in N glycosylation.

It has been reported that defects in N glycosylation lead to an increased sensitivity to hygromycin B and an increased resistance to orthovanadate (5, 13). To test if AGS1 is involved in N glycosylation, we examined the N glycosylation status of two secreted proteins, CPY and invertase (Suc2p), in ags1Δ mutants. N-glycosyl chains in CPY are short, consisting essentially of the core moiety of the dolichol-linked precursor structure; mutations in the core region lead to inefficient transfer of glycosyl chains from the dolichol precursor to CPY (4, 48). On the other hand, invertase N-glycosyl chains are extended to contain heterogeneous lengths of outer chains (3, 32).

The electrophoretic migration of CPY in ags1 mutants was not different from that of CPY produced by wild-type strains (Fig. 6A), indicating that AGS1 is not involved in the biosynthesis and transfer to the CPY protein of the dolichol-linked N-glycosyl precursor structure. As expected, underglycosylated CPY appeared in an alg5 mutant, which synthesizes a defective core structure (21).

FIG. 6.

FIG. 6

Open in a new tab

CPY and invertase (SUC) in yeast strains. (A) CPY in cellular proteins separated by SDS-PAGE was detected by immunoblotting with an anti-CPY antibody. The positions of mature CPY (mCPY) and underglycosylated CPY forms (1 and 2) are indicated. Lane 1, W10-2A (ags1); lane 2, W10-2B (ags1); lane 3, W10-2C (AGS1); lane 4, W10-2D (AGS1); lane 5, G2-10 (GDA2); lane 6, G2-11 (gda2); lane 7, PRY98 (alg5). (B) Invertase in cellular proteins was separated by nondenaturing gel electrophoresis followed by staining of its activity in the gel. Lanes are as in panel A. (C) Invertase in cellular proteins separated by SDS-PAGE was detected by immunoblotting with an anti-invertase antibody. The migration of invertase in the wild-type strain W10-5A (lane 2) is indicated (mSUC). Lane 1, W10-5C (ags1); lane 3, G2-10 (GDA2); lane 4, G2-11 (gda2). The migration of molecular mass markers (84 and 116 kDa) is indicated, and the expected migration of core-glycosylated invertase (about 90 kDa) is marked by an asterisk.

A defect in the addition of outer chains to the N-glycosyl core structure can be easily monitored by activity staining of invertase which has been separated on native polyacrylamide gels (32). Using this system, we found that invertase of ags1 mutants migrated significantly faster (Fig. 6B, lanes 1 and 2) than invertase of wild-type strains (Fig. 6B, lanes 3 and 4). Abnormal migration of invertase is more pronounced than in gda2 mutants (1), which have a partial defect in outer chain addition. These results could be confirmed by immunoblotting, by which denatured cellular proteins separated by SDS-PAGE were analyzed with an anti-invertase antibody. As expected, the bulk of invertase appeared as a heterogeneous smear around 150 kDa in wild-type cells (Fig. 6C, lane 2). In contrast, most invertase in the ags1 mutant produced distinct smaller bands (Fig. 6C, lane 1), with a major invertase species at 90 kDa, which is the size of core-glycosylated invertase (35). Again, the gda2 defect, although detectable in this system (Fig. 6C, lane 4), did not result in a phenotype as significant as observed for the ags1 mutant. Thus, it appears that the extension of N-glycosyl chains beyond the core structure is defective in ags1 mutants. No defects in N glycosylation of invertase and CPY were observed in a pho80 strain (VG51) (data not shown), suggesting that aminoglycoside supersensitivities in ags1 and pho80 strains are caused by different mechanisms.

Defective N glycosylation leads not only to supersensitivity to hygromycin B but also to resistance to orthovanadate (5, 32). To determine if ags1 mutants show this phenotype, we grew AGS1 wild-type strains (W10-2C and W10-2D) and ags1 mutant strains (W10-2A and W10-2B) on YPD agar containing 10 mM orthovanadate. While both wild-type strains were unable to grow, the ags1 mutants developed visible colonies after 2 days of growth (data not shown). Thus, ags1 mutants are vanadate resistant, as expected for a defect in N glycosylation.

Chitinase secreted by S. cerevisiae is highly O glycosylated (35). Neither ags1 nor pho80 mutants showed different electrophoretic mobilities of chitinase, indicating that defects in both genes do not perturb O glycosylation (data not shown).

DISCUSSION

An analysis of mutant strain RC1707 (15) revealed two genes, AGS1 and AGS3 (PHO80), whose expression is required to maintain basal levels of resistance to aminoglycoside antibiotics in yeast. The susceptibility to aminoglycoside antibiotics varies greatly among subtypes and species of yeasts, ranging from extreme sensitivity, as in S. cerevisiae RC1707, to high resistance, as in the human fungal pathogen Candida albicans. The reasons for these differences in aminoglycoside sensitivities are not known. Elucidation of resistance mechanisms may guide the development of antifungal agents; furthermore, sensitive strains may serve as improved host strains, since plasmids carrying genes encoding aminoglycoside-modifying enzymes, such as the phosphotransferases encoded by aph and hph, can be selected at low concentrations of antibiotics.

AGS1 was found to encode a small protein of 88 amino acids which had not been assigned a function previously (and which because of its small size is not a subject of a systematic functional analysis project [34]). Our evidence indicates that AGS1 is mutated in strain RC1707, since (i) the AGS1 transcript in RC1707 is shortened, indicating an extensive deletion; (ii) introduction of the wild-type AGS1 gene into RC1707 leads to the reappearance of the wild-type AGS1 transcript and restores hygromycin B and G418 resistance; and (iii) deletion of AGS1 in a wild-type strain leads to hygromycin B supersensitivity. Further analyses of the ags1Δ strain revealed that besides increased susceptibility to hygromycin B, it showed a higher resistance to vanadate and underglycosylated invertase, indicating a defect in N glycosylation. It has been reported previously that defects in N glycosylation lead to increased hygromycin B sensitivity and vanadate resistance; this phenotype is observed with mutants defective in the biosynthesis of the core region or outer N-glycosyl chains (5, 13). ags1 mutants did not show defects in growth or microscopic appearance, indicating that general cellular functions are not perturbed by the lack of Ags1p function. Particularly, growth on sucrose as the sole carbon source was not affected in ags1 strains, suggesting that underglycosylation of invertase is not due to a specific defect in invertase secretion. Thus, AGS1 appears to encode a hitherto undescribed component of the N glycosylation machinery. To explore whether core or outer chain biosynthesis is affected in ags1Δ mutants, we examined the glycosylation status of CPY, which only contains unextended, core-sized N-glycosyl chains (4). The ags1Δ mutant did not show any defect in CPY glycosylation, indicating that neither the activity of oligosaccharyltransferase nor the biosynthesis of the core region is affected by the Ags1 protein. In contrast, glycosylation of invertase, whose N-glycosyl chains are composed of core and outer chain regions (3, 35), was severely defective in ags1 mutants, leading to the accumulation of an invertase species with migration identical to that of the core-glycosylated form. Thus, it appears that a biosynthetic step in the elaboration of outer chains requires the Ags1 protein. Possibly, the ags1 mutation may lead to defects similar to those in mnn9 and och1 mutants, which are also defective in outer chain biosynthesis (3, 32). Because Ags1p, although hydrophobic, does not contain a signal sequence or a transmembrane region, it may influence N glycosylation in Golgi vesicles from their cytoplasmic sides, e.g., by interaction with the Och1p mannosyltransferase (32). Remarkably, a small hydrophobic protein is also required for the activity of the oligosaccharyltransferase complex in the endoplasmic reticulum (9). The molecular mechanism by which defects in N glycosylation lead to aminoglycoside sensitivity is not known. It appears that fully extended N-glycosyl chains of one or more proteins are required for basal levels of aminoglycoside resistance, possibly by permitting a high level of export or by preventing easy import of aminoglycosides. ags1 mutants may be useful host strains for the production of heterologous pharmaceutical glycoproteins, because the lack of extended N-glycosyl chains may increase specific activities and reduce antigenic properties of proteins secreted by yeast (11).

Surprisingly, AGS3 was found to be identical to the PHO80 gene, which is known to encode a protein acting as a cyclin to activate the Pho85p protein kinase (29, 30). The Pho80p-Pho85p complex is known to phosphorylate and thereby inactivate the Pho4p transcriptional activator, which is required to activate expression of PHO5, which encodes acid phosphatase (reviewed in references 24 and 36). Since the Pho80p-Pho85p complex is active only under high-phosphate conditions, PHO5 is expressed only in media containing low concentrations of phosphate. We found that both pho80 and pho85 mutants were supersensitive to aminoglycosides but that simultaneous mutation of PHO4 increased resistance levels. Thus, it appears that the supersensitivity phenotype requires an elevated activity (lack of phosphorylation) of the Pho4p protein; conversely, under high-phosphate conditions, Pho80p and Pho85p prevent activation of Pho4p and prevent aminoglycoside supersensitivity. The molecular mechanism by which activation of Pho4p leads to enhanced sensitivity is open to speculation. It is difficult to envisage a function of the PHO5-encoded acid phosphatase in this process, because Pho5p is secreted into the periplasm. A complex scenario in which aminoglycosides are phosphorylated in the cytoplasm, transported in the periplasm, and then dephosphorylated (activated) by acid phosphatase cannot be excluded at present and needs to be verified experimentally. Other explanations may be related to recent findings which have demonstrated that Pho proteins are involved in processes other than the biosynthesis of phosphatase. Pho85p is able to associate with cyclins other than Pho80p to function as a cyclin-dependent kinase, e.g., during the cell cycle (30). Since we observed that pho85 mutants are more sensitive to hygromycin B than pho80 mutants, it is possible that an unknown cyclin is partially redundant with Pho80p. The Pho80p-Pho85p kinase is known to phosphorylate targets other than Pho4p, e.g., enzymes involved in glycogen metabolism (20, 45). Also, it has been known for a long time that, in contrast to wild-type cells, pho80 (tup7) mutants are able to utilize 5′-mononucleotides (6), but the molecular mechanisms of this phenotype are unknown. Our finding that the pho4 mutation only partially restores resistance to hygromycin B in a pho85 strain could be due to an additional function of Pho85p with regard to aminoglycoside susceptibility that is independent of Pho4p.

ACKNOWLEDGMENTS

We are grateful to S. Harashima, B. Andrews, C. Abeijon, and F. Hilger for providing strains.

REFERENCES

  • 1.Abeijon C, Yanagisawa K, Mandon E C, Häusler A, Moremen K, Hischberg C B, Robbins P W. Guanosine diphosphatase is required for protein and sphingolipid glycosylation in the Golgi lumen of Saccharomyces cerevisiae. J Cell Biol. 1993;122:307–323. doi: 10.1083/jcb.122.2.307. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Adoutte-Panvier A, Davies J E. Studies of ribosomes of yeast species: susceptibility to inhibitors of protein synthesis in vivo and in vitro. Mol Gen Genet. 1984;194:310–317. [Google Scholar]
  • 3.Ballou C. Isolation, characterization and properties of Saccharomyces cerevisiae mnn mutants with nonconditional protein glycosylation defects. Methods Enzymol. 1990;185:440–470. doi: 10.1016/0076-6879(90)85038-p. [DOI] [PubMed] [Google Scholar]
  • 4.Ballou L, Hernandez L M, Alvarado E, Ballou C E. Revision of the oligosaccharide structures of yeast carboxypeptidase Y. Proc Natl Acad Sci USA. 1990;87:3368–3372. doi: 10.1073/pnas.87.9.3368. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Ballou L, Hitzeman R A, Lewis M S, Ballou C E. Vanadate-resistant yeast mutants are defective in protein glycosylation. Proc Natl Acad Sci USA. 1991;88:3209–3212. doi: 10.1073/pnas.88.8.3209. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Bisson L F, Thorner J. Mutations in the PHO80 gene confer permeability to 5′-mononucleotides in Saccharomyces cerevisiae. Genetics. 1982;102:341–359. doi: 10.1093/genetics/102.3.341. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Boles E, Göhlmann H W H, Zimmermann F K. Cloning of a second gene encoding 6-phosphofructo-2-kinase in yeast, and characterization of mutant strains without fructose-2,6-bisphosphate. Mol Microbiol. 1996;20:65–76. doi: 10.1111/j.1365-2958.1996.tb02489.x. [DOI] [PubMed] [Google Scholar]
  • 8.Chernoff Y O, Vincent A, Liebman S W. Mutations in eukaryotic 18S ribosomal RNA affect translational fidelity and resistance to aminoglycoside antibiotics. EMBO J. 1994;13:906–913. doi: 10.1002/j.1460-2075.1994.tb06334.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Chi J H, Roos J, Dean N. The OST4 gene of Saccharomyces cerevisiae encodes an unusually small protein required for normal levels of oligosaccharyltransferase activity. J Biol Chem. 1996;271:3132–3140. doi: 10.1074/jbc.271.6.3132. [DOI] [PubMed] [Google Scholar]
  • 10.Cui Y, Dinman J D, Peltz S W. mof4-1 is an allele of the UPF1/IFS2 gene which affects both mRNA turnover and −1 ribosomal frameshifting efficiency. EMBO J. 1996;15:5726–5736. doi: 10.1002/j.1460-2075.1996.tb00956.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Cumming D A. Glycosylation of recombinant protein therapeutics: control and functional implications. Glycobiology. 1991;1:115–130. doi: 10.1093/glycob/1.2.115. [DOI] [PubMed] [Google Scholar]
  • 12.Davies J E. Resistance to amonoglycosides: mechanisms and frequency. Rev Infect Dis. 1983;5:S261–S266. [Google Scholar]
  • 13.Dean N. Yeast glycosylation mutants are sensitive to aminoglycosides. Proc Natl Acad Sci USA. 1995;92:1287–1291. doi: 10.1073/pnas.92.5.1287. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.De Stasio E A, Moazed D, Noller H F, Dahlberg A E. Mutations in 16S ribosomal RNA disrupt antibiotic-RNA interactions. EMBO J. 1989;8:1213–1216. doi: 10.1002/j.1460-2075.1989.tb03494.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Ernst J F, Chan R K. Characterization of Saccharomyces cerevisiae mutants supersensitive to aminoglycoside antibiotics. J Bacteriol. 1985;163:8–14. doi: 10.1128/jb.163.1.8-14.1985. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Gietz R D, Sugino A. New yeast-Escherichia coli shuttle vectors constructed with in vitro mutagenized yeast genes lacking six-base restriction sites. Genes. 1988;74:527–534. doi: 10.1016/0378-1119(88)90185-0. [DOI] [PubMed] [Google Scholar]
  • 17.Gilliquet V, Legrain M, Berben G, Hilger F. Negative regulatory elements of the Saccharomyces cerevisiae PHO system: interaction between PHO80 and PHO85 proteins. Gene. 1990;96:181–188. doi: 10.1016/0378-1119(90)90251-l. [DOI] [PubMed] [Google Scholar]
  • 18.Gritz L, Davies J. Plasmid-encoded hygromycin B phosphotransferase gene and its expression in Escherichia coli and Saccharomyces cerevisiae. Gene. 1983;25:179–188. doi: 10.1016/0378-1119(83)90223-8. [DOI] [PubMed] [Google Scholar]
  • 19.Güldener U, Heck S, Fiedler T, Beinhauer J, Hegemann J H. A new efficient gene disruption cassette for repeated use in budding yeast. Nucleic Acids Res. 1996;24:2519–2524. doi: 10.1093/nar/24.13.2519. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Huang D, Farkas I, Roach P J. Pho85p, a cyclin-dependent protein kinase, and the Snf1p protein kinase act antagonistically to control glycogen accumulation in Saccharomyces cerevisiae. Mol Cell Biol. 1996;16:4357–4365. doi: 10.1128/mcb.16.8.4357. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Huffaker T C, Robbins P W. Yeast mutants deficient in protein glycosylation. Proc Natl Acad Sci USA. 1983;80:7466–7470. doi: 10.1073/pnas.80.24.7466. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Ito H, Fukuda Y, Murata K, Kimura A. Transformation of intact yeast cells treated with alkali cations. J Bacteriol. 1983;153:163–168. doi: 10.1128/jb.153.1.163-168.1983. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Jones E. Vacuolar proteases in yeast Saccharomyces cerevisiae. Methods Enzymol. 1990;185:372–386. doi: 10.1016/0076-6879(90)85033-k. [DOI] [PubMed] [Google Scholar]
  • 24.Lenburg M E, O’Shea E K. Signaling phosphate starvation. Trends Biochem Sci. 1996;21:383–387. [PubMed] [Google Scholar]
  • 25.Li M, Tzagoloff A, Underbrink-Lyon K, Martin N C. Identification of the paromomycin-resistance mutation in the 15S rRNA gene of yeast mitochondria. J Biol Chem. 1982;257:5921–5928. [PubMed] [Google Scholar]
  • 26.Liebman S W, Cavenagh M. An antisuppressor that acts on omnipotent suppressors in yeast. Genetics. 1980;95:49–61. doi: 10.1093/genetics/95.1.49. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Madden S L, Creasy C L, Srinivas V, Fawcett W, Bergman L W. Structure and expression of the PHO80 gene of Saccharomyces cerevisiae. Nucleic Acids Res. 1988;16:2625–2637. doi: 10.1093/nar/16.6.2625. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.McCusker J H, Haber J E. crl mutants of Saccharomyces cerevisiae resemble both mutants affecting general control of amino acid biosynthesis and omnipotent translational suppressor mutants. Genetics. 1988;119:317–327. doi: 10.1093/genetics/119.2.317. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Measday V, Moore L, Ogas J, Tyers M, Andrews B. The PCL2 (ORFD)-PHO85 cyclin-dependent kinase complex: a cell cycle regulator in yeast. Science. 1994;266:1391–1395. doi: 10.1126/science.7973731. [DOI] [PubMed] [Google Scholar]
  • 30.Measday V, Moore L, Retnakaran R, Lee J, Donovell M, Neiman A M, Andrews B. A family of cyclin-related proteins that interact with the Pho85 cyclin-dependent kinase. Mol Cell Biol. 1997;17:1212–1223. doi: 10.1128/mcb.17.3.1212. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Moazed D, Noller H F. Interaction of antibiotics with functional sites in 16S ribosomal RNA. Nature. 1987;327:389–394. doi: 10.1038/327389a0. [DOI] [PubMed] [Google Scholar]
  • 32.Nakayama K, Nagasu T, Shimma Y, Kuromitsu J, Jigami Y. OCH1 encodes a novel membrane bound mannosyltransferase: outer chain elongation of asparagine-linked oligosaccharides. EMBO J. 1992;11:2511–2519. doi: 10.1002/j.1460-2075.1992.tb05316.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Nasmyth K, Reed S. Isolation of genes by complementation in yeast: molecular cloning of a cell-cycle gene. Proc Natl Acad Sci USA. 1980;77:2119–2123. doi: 10.1073/pnas.77.4.2119. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Oliver S G. From gene to screen with yeast. Curr Opin Genet Dev. 1997;7:405–409. doi: 10.1016/s0959-437x(97)80156-6. [DOI] [PubMed] [Google Scholar]
  • 35.Orlean P, Kuranda M J, Albright C F. Analysis of glycoproteins from Saccharomyces cerevisiae. Methods Enzymol. 1991;194:682–697. doi: 10.1016/0076-6879(91)94050-m. [DOI] [PubMed] [Google Scholar]
  • 36.Oshima Y, Ogawa N, Harashima S. Regulation of phosphatase synthesis in Saccharomyces cerevisiae—a review. Gene. 1996;179:171–177. doi: 10.1016/s0378-1119(96)00425-8. [DOI] [PubMed] [Google Scholar]
  • 37.Palmer E, Wilhelm J M, Sherman F. Phenotypic suppression of nonsense mutants in yeast by aminoglycoside antibiotics. Nature. 1979;277:148–150. doi: 10.1038/277148a0. [DOI] [PubMed] [Google Scholar]
  • 38.Perlin D S, Brown C L, Haber J E. Membrane potential defect in hygromycin B-resistant pma1 mutants of Saccharomyces cerevisiae. J Biol Chem. 1988;263:18118–18122. [PubMed] [Google Scholar]
  • 39.Prezant T R, Chaltraw W E, Jr, Fischel-Ghodsian N. Identification of an overexpressed yeast gene which prevents aminoglycoside toxicity. Microbiology. 1996;142:3407–3414. doi: 10.1099/13500872-142-12-3407. [DOI] [PubMed] [Google Scholar]
  • 40.Rank G H, Robertson A J, Phillips K L. Modification and inheritance of pleiotropic cross resistance and collateral sensitivity in Saccharomyces cerevisiae. Genetics. 1974;80:483–493. [PMC free article] [PubMed] [Google Scholar]
  • 41.Rose M, Novick P. A Saccharomyces cerevisiae plasmid bank based on a centromere-containing shuttle vector. Gene. 1987;60:237–243. doi: 10.1016/0378-1119(87)90232-0. [DOI] [PubMed] [Google Scholar]
  • 42.Schmitt M E, Brown T A, Trumpower B L. A rapid and simple method for preparation of RNA from Saccharomyces cerevisiae. Nucleic Acids Res. 1990;18:3091–3092. doi: 10.1093/nar/18.10.3091. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Sherman F, Fink G, Hicks J. Methods in yeast genetics. Cold Spring Harbor, N.Y: Cold Spring Harbor Laboratory; 1981. [Google Scholar]
  • 44.Singh A, Ursic D, Davies J. Phenotypic suppression and misreading in Saccharomyces cerevisiae. Nature. 1979;277:146–148. doi: 10.1038/277146a0. [DOI] [PubMed] [Google Scholar]
  • 45.Timblin B K, Tatchell K, Bergman L W. Deletion of the gene encoding the cyclin-dependent protein kinase Pho85 alters glycogen metabolism in Saccharomyces cerevisiae. Genetics. 1996;143:57–66. doi: 10.1093/genetics/143.1.57. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45a.Timpel, C., and J. F. Ernst. Unpublished results.
  • 46.Von Aasen U, Noller H F. Footprinting the sites of interaction of antibiotics with catalytic group I intron RNA. Science. 1993;260:1500–1503. doi: 10.1126/science.8502993. [DOI] [PubMed] [Google Scholar]
  • 47.Zapp M L, Stern S, Green M R. Small molecules that selectively block RNA binding of HIV-1 Rev protein inhibit Rev function and viral production. Cell. 1993;74:969–978. doi: 10.1016/0092-8674(93)90720-b. [DOI] [PubMed] [Google Scholar]
  • 48.Zufferey R, Knauer R, Burda P, Stagljar I, te Heesen S, Lehle L, Aebi M. STT3, a highly conserved protein required for yeast oligosaccharyl transferase in vivo. EMBO J. 1995;14:4949–4960. doi: 10.1002/j.1460-2075.1995.tb00178.x. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Journal of Bacteriology are provided here courtesy of American Society for Microbiology (ASM)

RESOURCES